CN111316115B

CN111316115B - Method for detecting self-discharge defects in battery cells

Info

Publication number: CN111316115B
Application number: CN201880048454.2A
Authority: CN
Inventors: M·门斯勒; B·德洛贝尔; A·L·德里梅耶弗兰科; B·阿尔萨克; M·奥里古奇
Original assignee: Renault SAS
Current assignee: Renault SAS
Priority date: 2017-07-19
Filing date: 2018-07-17
Publication date: 2022-06-24
Anticipated expiration: 2038-07-17
Also published as: EP3655788A1; CN111316115A; JP7166298B2; US20200166582A1; FR3069180A1; KR20200032124A; EP3655788B1; FR3069180B1; WO2019016218A1; JP2020527697A; KR102543469B1

Abstract

The method is used for detecting a self-discharge defect in a cell (1, 2, 3, 4) in an electrical storage battery (10) having a plurality of battery cells (1, 2, 3, 4), wherein: at least partially charge balancing the battery cells (1, 2, 3, 4); -relaxation of the battery cells (1, 2, 3, 4); for each battery cell (i), calculating the charge balance (Σ i) of said cell (i) during this balancing and relaxation; and for each battery cell (i), detecting any self-discharge defects of said cell (i) as a function of the charge balance (Σ i) calculated for said cell (i) during this balancing and relaxation.

Description

Method for detecting self-discharge defects in battery cells

Technical Field

The invention relates to the control of an electric storage battery having a plurality of cells, which is intended in particular to be integrated into a motor vehicle.

Background

Electrically propelled motor vehicles include in particular purely electrically propelled vehicles, hybrid vehicles or rechargeable hybrid vehicles. Such vehicles are equipped with an electric storage battery having a large number of battery cells. These units may be mounted in series and/or in parallel.

As a general rule, these cells forming the electric storage battery have characteristics similar to each other. However, these cells may be subject to deviation or variation over the life of the battery. For example, there may be variations in cell capacity, variations in cell resistance, or variations in cell charge state or variations in cell temperature may temporarily exist. These deviations can lead to different degrees of aging of each cell of the storage battery and to deviations in the state of health of these cells.

The electric storage cells are directly affected by deviations with respect to their constituent units as a whole, and in particular by deviations in the state of charge. In fact, as the charge difference between these cells increases, the total available capacity of the battery decreases.

A common way to overcome this problem is to regularly balance the units. This balancing may be done directly by the battery management system operating in an autonomous manner. For example, cell balancing may be dissipative in that the charge state of the cell is balanced by discharging the cell pair resistance towards the target charge state. By means of this solution, the deviation of the load state can be kept below a threshold value, thereby limiting the drop in the total available capacity of the battery.

However, this solution is not entirely satisfactory. In fact, the cells of the storage batteries may encounter internal problems that lead to a rapid increase in self-discharge over time. Such self-discharge may increase state of charge deviation between the battery cells. If the self-discharge becomes too large, the balancing system loses its effectiveness and eventually becomes unable to compensate for the state of charge deviation between the cells of the storage battery. This often leads to a rapid and considerable drop in the capacity of the storage cell, or even to a cell breakdown, resulting in a breakdown of the entire storage cell.

Document CN 105527583 describes a method of detecting self-discharge in a battery pack, in which a self-discharge defect of a battery cell is detected in the event that the voltage at the terminals of the cell varies significantly between two predetermined instants. While such an approach may enable detection of self-discharge defects in the cells, there may still be self-discharge defects that are not detected and/or false self-discharge defect warnings issued. Therefore, there is a need to provide a more reliable method for detecting self-discharge defects.

Furthermore, the method described in CN 105527583 requires a longer relaxation time of the battery between the two voltage measurement instants. Thus, the method can only be used during long periods of vehicle inactivity, for example when the vehicle remains in a parking space. Such pauses occur infrequently. The detection of self-discharge defects is thus delayed or even impossible.

Disclosure of Invention

In view of the above, it is an object of the present invention to detect a self-discharge defect of a cell of an electric storage battery at an early stage so as to avoid a state of charge deviation between cells.

For this purpose, a method for detecting a self-discharge defect in a cell in an electrical storage battery having a plurality of battery cells is proposed, wherein:

-balancing the charge of the battery cells at least partially,

-relaxation of the battery cells,

-for each battery cell, calculating the charge balance of said cell during this balancing and relaxation, and

-for each battery cell, detecting the possible presence of any self-discharge defect of said cell based on the charge balance during the balancing and relaxation calculated for said cell. Therefore, the self-discharge defect of the cell is detected by analyzing the effectiveness of the balancing of the battery cell. This makes the detection of any self-discharge defects in each cell of this battery more reliable.

Advantageously, for each battery cell, the charge balance of said cell during the balancing and relaxation is calculated taking into account at least one quantity chosen from: the charge to be balanced for the cell immediately before the cell is initially balanced, the charge to be balanced for the cell immediately after the cell is relaxed, and an amount of charge that the cell discharges during the balancing.

Preferably, for any battery cell, the charge balance of the cell during the balancing and relaxation is calculated using the following relationship:

∑_i＝q_{i, balance}(t₂)+Δq_i(T)-q_{i, balance}(t₁)

Wherein, "∑_i"denotes the charge balance of the cell during the equilibrium and relaxation,

“q_{i, balance}(t) "represents the charge to be balanced for the cell at time t,

“Δq_i(T) "represents the amount of charge discharged by the cell during time interval T,

“t₁"denotes the moment immediately before the start of balancing the cell,

“t₂"denotes a time immediately after the end of relaxation of the unit, and

"T" denotes a period during which balancing has been performed on the charge of the cell.

In one embodiment, the balancing comprises, for each cell: a step of calculating a charge to be balanced for the cell; and a step of performing the balancing, in which an amount of electric charge discharged by the cell is calculated; and this balancing step continues as long as the charge to be balanced for the cell is strictly greater than the amount of charge discharged by the cell.

This embodiment, in which the expected charge to be balanced is compared with the amount of actually discharged charge, allows a particularly accurate balancing. Therefore, the accuracy of detecting the self-discharge defect is improved.

In another embodiment, the balancing comprises, for each cell: a step of calculating a charge to be balanced for the cell; and a step of calculating a balancing time of the cell based on the calculated electric charges to be balanced for the cell, wherein the balancing is performed during the calculated balancing time.

This embodiment based on the calculation of the expected equilibration time requires relatively less computational resources to detect self-discharge defects.

Advantageously, for each cell, the charge to be balanced for the cell is calculated based on at least one parameter selected from: a state of charge of the cell, a state of charge of a target cell, a capacity of the cell, a capacity of a target cell, a state of health of the cell, a state of health of a target cell, a zero current voltage of the cell, a zero current voltage of a target cell, a nominal capacity of the cell, and a nominal capacity of a target cell.

Preferably, for each cell, the amount of charge that the cell discharges during a certain period is calculated based on at least one parameter selected from: a balancing resistance, a voltage at a terminal of the cell at a time within the period, a balancing current of a cell of the electrical storage cell, and a nominal voltage of a cell of the electrical storage cell.

In one embodiment, the presence of a self-discharge defect in the first battery cell is detected when: for any second battery cell other than the first battery cell, the charge to be balanced by the second cell immediately after the relaxation of the second cell ends is strictly greater than the charge to be balanced by the second cell immediately before the balancing of the second cell begins, and the amount of charge discharged by the second cell during the balancing and relaxation is not zero.

Such an embodiment enables detection of a self-discharge defect in a battery cell whose charge state corresponds to that of the target cell.

For each cell, it may also be provided that the presence of a self-discharge defect in the cell is detected in the following cases: during this balancing and relaxation the calculated charge balance of the cell exceeds a strictly positive threshold, said threshold being determined based on at least one error selected from: an error in the amount of charge to be balanced for the cell immediately before the balancing of the cell is started, an error in the amount of charge to be balanced for the cell immediately after the relaxation of the cell is ended, and an error in the amount of charge discharged by the cell during the balancing and the relaxation.

By applying such a threshold value, it is possible to take into account the sources of inaccuracy associated with the storage battery, which could prevent detection or, conversely, could lead to erroneous detection.

Advantageously, for any cell, the threshold is determined based on an error for the charge to be balanced for the cell, the error being calculated for any instant in time according to the following expression:

δq_{i, balance}＝|Q_i.δSOC_i|+|SOC_i.δQ_i|+|Q_Target.δSOC_Target|+|SOC_Target.δQ_Target|

Wherein, the value of "δ q_{i, balance}"denotes the error in the charge to be balanced by the cell at that moment,

“δSOC_i"represents the error in the charge state of the cell,

“δQ_i"indicates an error in the charge of the cell,

“SOC_i"represents the charge state of the cell at that time,

“Q_i"denotes the charge of the cell at the time,

“δSOC_target"represents the error in the charge state of the target cell,

“δQ_target"indicates an error in the charge of the target cell,

“SOC_target"represents the charge state of the target cell at that time, and

“Q_target"indicates the charge of the target cell at the time.

Preferably, for any cell, the threshold is determined based on an error in the amount of charge that the cell discharges during a certain time interval, the error being calculated using the following expression:

wherein "T" represents the time interval,

“δΔq_i(T) "represents an error in the amount of charge discharged by the cell during the time interval,

“R_balancing"denotes the resistance for this balance, and

for any time instant τ, V, belonging to said time interval_i(τ) "represents the voltage measured at the terminals of the cell at that instant τ.

For any cell, the determination of the threshold may also be provided based on an error in the amount of charge that the cell discharges during a certain time interval, the error being calculated using the following expression:

δΔq_i(T)＝|T.δI_balancing|

Wherein "T" represents the time interval,

“δΔq_i(T) "represents an error in the amount of charge discharged by the cell during the time interval, and

“δI_balancing"indicates an error in the balance current of the cells of the storage battery.

As explained below, these error calculations take into account most of the inaccuracies encountered in the storage battery. In this way, the calculation of the threshold is refined, and thus the reliability of detecting self-discharge defects is improved.

Drawings

Other objects, characteristics and advantages of the present invention will appear from the following description, provided purely by way of non-limiting example and with reference to the accompanying drawings, in which:

figure 1 schematically shows an electric storage cell,

fig. 2 schematically shows a detection method according to an exemplary embodiment of the invention;

FIG. 3 shows three graphs illustrating the calculation of the charge to be balanced for the cells of the battery of FIG. 1,

fig. 4 shows a graph demonstrating the charge balance of the cell of the battery of fig. 1 during balancing and relaxation, according to a first operating condition, and

fig. 5 shows a graph illustrating the charge balance of the cell of the battery of fig. 1 during balancing and relaxation, according to a second operating condition.

Detailed Description

Referring to fig. 1, an electricity storage cell 10 is schematically shown. The battery 10 is designed to be incorporated into an electrically propelled motor vehicle for powering an electric traction machine belonging to the powertrain of the motor vehicle.

The battery 10 includes two terminals 14. Battery 10 includes four battery cells (denoted 1, 2, 3, and 4, respectively). In the illustrated example,

units

1, 2, 3 and 4 are connected in series. However, it should be possible to connect the cells in parallel, or alternatively to connect some cells in parallel and the rest in series, without departing from the scope of the invention. Similarly, it is obviously possible to have a different number of units without departing from the scope of the invention.

The battery 10 has an information connection to a management system 12. The management system 12 is also called the english expression "battery management system" or "BMS" for short. The system 12 manages the various methods used during the life of the battery 10. For example, as explained below, the management system 12 manages a method of balancing the cells of the battery 10.

Each of the

cells

1, 2, 3, 4 is associated with a corresponding balancing circuit 16. In the illustrated example, balancing circuit 16 is the same for all cells of battery 10. For any

cell

1, 2, 3 or 4, the balancing circuit 16 associated with that cell includes a circuit 18 connected to both terminals of that cell. In the circuit 18, a switch 20 is connected in series with a balancing resistor 22. However, the present invention is not limited to this form of balancing circuit. In particular, it should also be possible to make a single balancing circuit common to all the cells of the battery 10 and to be able to perform balancing of any one of these cells separately.

The system 12 is provided with hardware means and software means for receiving data related to the

units

1, 2, 3 and 4, as indicated by the dashed arrow 23. The system 12 is provided with hardware means and software means for determining intermediate data related to the

units

1, 2, 3 and 4 based on the received data.

For example, for any cell i selected from

cells

1, 2, 3, and 4, during balancing of this cell i, system 12 can receive the voltage v at the terminals of cell i_iAnd a current I flowing through the cell I_{i, balance}. System 12 is capable of determining the zero-current voltage OCV of cell i_i(also known by the english expression "open circuit voltage") and the state of charge SOC of cell i_i(also called the english expression "state of charge)").

As indicated by the dashed arrow 24, the system 12 is provided with hardware and software means for opening and closing the switches 20 associated with the

units

1, 2, 3 and 4, respectively. Thus, for any cell i (where i is in the range from 1 to 4), if the system 12 deems it necessary to balance the cell i, it closes the switch 20 of the balancing circuit 16 corresponding to the cell i. The electrical energy discharged by cell i is dissipated in the balancing resistor 22 of the balancing circuit 16 associated with cell i. If the system 12 deems it necessary to stop balancing the cell i, the system opens the switch 20 of the balancing circuit 16 associated with the cell i.

The execution of the method of detecting self-discharge defects according to the present invention will now be described with reference to fig. 2 to 5. The method is performed in order to detect self-discharge defects in the

cells

1, 2, 3 and 4 of the battery 10 using the management system 12. An example of performing the method according to the invention is schematically illustrated in fig. 2.

The method according to the invention comprises a first phase P01 of balancing the charges of the cells of the battery 10. In describing phase P01, reference may be made to FIG. 3, which illustrates the capacity Q of each of

units

1, 2, 3, and 4₁、Q₂、Q₃And Q₄A first graph 26 of the distribution of (a). These capacities are shown in a gaussian distribution. The second graph 28 contained in fig. 3 shows the electricity with the

cells

1, 2, 3 and 4The relationship between the zero-current voltage OCV of the cell that varies according to the state of charge SOC.

Phase P01 includes a first step E01, in which target parameters are determined. More specifically, a target state of charge SOC is determined_TargetAnd target capacity Q_Target. In the illustrated example, the target state of charge SOC_TargetAnd target capacity Q_TargetState of charge SOC defined as the cell i with the lowest state of charge_iAnd capacity Q_i. In the case of FIG. 3, the target state of charge SOC_TargetCorresponding to the state of charge SOC of the cell 4₄And target capacity Q_TargetIs defined to be equal to the capacity Q of the cell 4₄：

Phase P01 includes a second step E02 in which the expected charge to be balanced for the cells of battery 10 is calculated. For any cell i, the expected charge q to be balanced for cell i_{i, balance}Corresponds to the charge that must be discharged into the resistor 22 of the balancing circuit 16 associated with cell i in order to balance cell i with respect to the target cell. For this purpose, the following equation may be applied:

q_{i, balance}＝SOC_i×Q_i-SOC_Target×Q_Target (2)

In the illustrated example, it is advantageously assumed that there is no deviation between the capacities of the cells of the battery 10. Thus, if Q represents the capacity of all cells, the charge Q can be calculated by applying the following equation_{i, balance}：

q_{i, balance of}＝(SOC_i-SOC_Target)×Q (3)

The method according to the invention is not limited to the calculations described above by way of example. In a first example of variant, for any cell i, the charge q may be calculated by applying the following equation, without departing from the scope of the invention_{i, balance of}：

q_{i, balance}＝f(OCV_i)×(SOH_i×Q_{i, nominal})-f(OCV_Target)×(SOH_Target×Q_{Target, nominal})

(4)

In this equation, the function f represents the function whose graphical representation corresponds to the graph 28 of fig. 3, SOH_iIndicates the health status of cell i (also known by the english term "health"), Q_{i, nominal}Indicating the nominal capacity, SOH, of the cell i_TargetRepresents a target state of health, and Q_{Target, nominal}Representing the target nominal capacity. For example, target State of health SOH_TargetAnd target nominal capacity Q_{Target, nominal}Respectively, the state of health and the nominal capacity of the target cell, in this case the state of health SOH of the cell 4₄And nominal capacity Q_{4, nominal rating}。

In particular, the equation according to the first example of variant can be used in the case where the management system 12 is provided with means for determining the state of health of each of the units 1 to 4 separately.

According to a second example of a variant of the method according to the invention, the determination of the target state of charge SOC may be selected_TargetIn different ways. For example, target State of Charge SOC_TargetMay be an average of the state of charge of the cells of the battery 10. This may result in an expected charge q of the cell i to be balanced_{i, balance}Taking a negative value. However, since the balancing system is dissipative, it is not possible to recharge the respective unit. Thus, in this case, for any cell i, when the expected charge q to be balanced is balanced_{i, balance}This charge is automatically defined as equal to zero when the calculation of (c) gives a negative result.

The expected charge q to be balanced for the

individual cells

1, 2, 3 and 4 is schematically shown on the third graph 48 of fig. 3_{1, balancing}、q_{2, balancing}、q_{3, balancing}And q is_{4, balancing}. Graph 48 shows, for each cell i selected from

cells

1, 2, 3 and 4, the charge q of said cell_iAnd the charge of the target cell (here)In this case q₄). For each cell i (where i is in the range of 1 to 4), the expected charge q to be balanced_{i, balance}Schematically indicated by a downwardly directed vertical arrow. As shown in graph 48, the expected charge q to be balanced for the target cell 4_{4, balancing}Equal to zero.

Phase P01 includes a third step E03, in which the balancing of the cells of the battery 10 is carried out. To this end, the system 12 closes the switch 20 associated with each cell to be balanced. At the same time, at each time t and for each cell i of battery 10, system 12 calculates the amount of charge Δ q actually discharged by cell i at time t_i(t) of (d). For any cell i of the battery 10, the amount of charge actually discharged by the cell i at time t may be calculated according to the following expression:

now, ohm's law states:

wherein R is_BalancingIs the value of resistor 22 of balancing circuit 16.

Thus, the quantity Δ q can be calculated by applying the following equation_i(t)：

The expected charge q that the system 12 will constantly balance for any cell i of the battery 10_{i, balance}Amount of charge Δ q actually discharged_i(t) comparison was performed. As long as the expected charge q is to be balanced_{i, balance}Strictly greater than the actual discharged charge Δ q_i(t), the switch 20 remains closed. If the charge q is_{i, balance}Becomes equal to or greater than the amount Δ q_i(t), the switch 20 associated with cell i is opened. Then, the pair unit is terminatedAnd i is balanced.

Due to the expected charge q to be balanced for cell 4_{4, balancing}Is zero, so at any time t, the cell actually discharges an amount of charge Δ q₄(t) cannot be strictly smaller than the expected charge to be balanced. Thus, in the example illustrated, during step E03, switch 20 associated with cell 4 is not closed.

According to a third example of a variant of the method according to the invention, the balancing may be controlled on the basis of the calculated balancing time. In step E02 of this example of a variant of the method, for any cell i of the battery 10, the expected charge q to be balanced for cell i is used_{i, balance}To calculate the equilibrium time t of the cell i_{i, balance}. This equilibrium time t for cell i is calculated by applying the following equation_{i, balance}：

Wherein,

wherein, V_{Nominal scale}Is the nominal voltage of the cells of the battery 10. In a third example of variant, it is assumed that the deviation of the cell's equilibrium resistance and the deviation of the nominal voltage are zero. Thus, assume a balanced current I_{i, balance}Is the same for all units and is denoted as I_Balancing。

In view of the above, for any cell i of battery 10, the equilibrium time t may be calculated by applying the following formula_{i, balance}：

In step E03 of the third example of variant, for any cell i of battery 10, at time t₁And time t₁+t_{i, balance}In between, the system 12 sets the switch 20 associated with cell i to a closed state, where t₁Is the moment when equilibrium begins to take place.

The third example of the variation is advantageous in that: it is not necessary to constantly calculate the quantity Δ q of all cells_i(t) of (d). Thus, the demand in terms of computing resources is reduced. However, the balance becomes more approximate because the cell's nominal voltage (which is constant) is used instead of the actual voltage (which varies during use of the battery 10) to calculate the balance time.

In a fourth example of the variant, the charge quantity Δ q actually discharged by the cell i at the instant t can be calculated in different ways_i(t) of (d). In this fourth example of variant, the quantity Δ q is calculated for any cell i by applying the following equation_i(t)：

Δq_i(T)＝I_Balancing×t_{i, balance} (11)

As a general rule, balancing is only possible during the mission of the vehicle (i.e. when the vehicle is running and under control). Thus, the balancing time is typically too short to allow all cells to be balanced in a single task. In this case, the balance is interrupted and resumed after a period of interruption.

In the exemplary embodiment illustrated, the target parameters are updated when equilibrium is restored after an interruption. This option is advantageous, especially because the cell may be in a high charge state when equilibrium is restored.

However, the present invention is not limited to this option, and it should be possible that there is a fifth example of a variation in which the target parameter is not updated when the balance is restored after the interruption. This fifth example is advantageous in that it requires less computing resources.

The detection method according to the invention then comprises, for each cell of the battery 10, a phase P02 of relaxation of this cell. During a phase P02, at a relaxation time T_rDuring which the current of the cell remains equal to zero.

In the exemplary embodiment illustrated, based on electricityEquilibration time t of the cells of the cell 10_{i, balance}To calculate the relaxation time T_r. More precisely, time T_rSubstantially equal to the equilibrium time t of the cells of the battery 10_{i, balance}Multiplied by a factor ranging from 0.5 to 1.

Fig. 4 shows the variation in time of the charge in cell i of battery 10 subjected to an equilibrium phase P01 and a relaxation phase P02 of the method according to the invention. Cell i (the charge variation in which is shown in fig. 4) does not experience any charge loss due to self-discharge defects.

Equilibrium at time t₁Starting and interrupting after the expiration of the equilibration period T. The period T may be less than or equal to the equilibrium time T calculated for the cell i_{i, balance}. If the balance is interrupted, unit i is relaxed until time t₂。

The square 32 schematically shows the cell i at the instant t₁Electric charge q of_i(t₁). Square 34 shows cell i at time t₂Electric charge q of_i(t₂). Horizontal line 36 represents the target unit at time t₁Electric charge q of_Target(t₁). Horizontal line 38 represents the target unit at time t₂Electric charge q of_Target(t₂). A first vertical downward arrow 40 indicates at a time t₁Estimated expected charge q to balance for cell i_{i, balance}(t₁). The second vertical downward arrow 42 represents the amount of charge Δ q actually discharged by cell i during equilibrium_i(T). A third vertical downward arrow 44 schematically shows at a time t₂Estimated expected charge q to be balanced for cell i_{i, balance}(t₂)。

As mentioned above, in the case of fig. 4, cell i does not exhibit any self-discharge defects. Thus, the data relating to the charge of cell i follows the equation:

q_{i, balance}(t₂)＝q_{i, balance of}(t₁)-Δq_i(T) (12)

Fig. 5 shows the change over time of the charge in the cell i which has undergone the same phases as the equilibrium phase P01 and the relaxation phase P02 in fig. 4. Cell i (the charge variation of which is shown in fig. 5) suffers charge loss due to significant self-discharge defects.

As in FIG. 4, the equilibrium is at time t₁Starting and interrupting after expiry of the equilibrium period T, then the unit i is relaxed until the instant T₂。

The graph of fig. 5 differs from the graph of fig. 4 in that: the fourth vertical downward arrow 46 schematically shows the amount of charge Δ q of the cell i discharged as a result of self-discharge during phases P01 and P02 of the method according to the invention_i,s(T+T_r)。

Since the cell i exhibits a self-discharge defect in the case of fig. 5, the data related to the charge of the cell cannot conform to equation (12). Instead, in the case of fig. 5, the data related to the charge of cell i follows the following equation:

q_{i, balance}(t₂)＝q_{i, balance}(t₁)-Δq_i(T)-Δq_i,s(T+T_r) (13)

In which the cell i exhibits a very large self-discharge defect, or in which the equilibrium time interval T and/or the relaxation time interval T_rIn the very extreme case, at time t₁And time t₂May be discharged with a greater amount of charge than at time t₁Calculated expected charge q to be discharged for cell i_{i, balance}(t₁)。

In this case, it is preferable that the air conditioner,

q_{i, balance of}(t₁)-Δq_i(T)-Δq_i,s(T+T_r)<0 (14)

In view of the above, for any unit i of battery 10, the charge balance Σ of unit i can be calculated by applying the following equation_iTo detect self-discharge defects of cell i:

∑_i＝q_{i, balance}(t₂)+Δq_i(T)-q_{i, balance of}(t₁) (15)

In the theoretical model, for any cell i of the battery 10, if the charge balance ∑ is_iZero, no self-discharge defect of cell i exists.

Conversely, for any cell i of the battery 10, if a strictly positive balance Σ has been calculated_iThen there is a self-discharge defect of cell i. In this case, the amount of charge Δ q of the cell i discharged due to self-discharge can be calculated by applying the following equation_i,s(T+T_r)：

q_i,s(T+T_r)＝q_{i, balance}(t₁)-q_{i, balance}(t₂)-q_i(T) (16)

The self-discharge current I of cell I can be estimated by applying the following equation_i,s：

The foregoing has shown that the charge balance sigma is calculated_iTo detect the presence of self-discharge defects in the cells i of the battery 10. However, this model is not applicable to target units. Another calculation for detecting self-discharge defects in a target cell will now be described in detail.

For such calculation, it is necessary to detect whether the target cell is discharged faster than the other cells of the battery 10 during the detection method. More specifically, for any cell i of battery 10, where cell i is a cell other than the target cell, if

q_i(t₂)>q_i(t₁) Wherein q is_i(T)≠0 (18)

A self-discharge defect is detected in the target cell.

Therefore, the theoretical model shown above can be used to detect self-discharge defects in any cell of the battery 10. In practice, however, a threshold must be applied to achieve effective detection in order to take into account all sources of inaccuracy that would prevent detection or otherwise lead to false detections.

In particular, for any cell i of the battery 10, it is calculated to be balanced for cell iPredicted charge q_{i, balance}Inaccuracies may occur. These inaccuracies may be caused by, among other things:

voltage sensor measuring zero current voltage OCV_i，

-when zero current voltage OCV is measured_iThe previously insufficient relaxation time of the ring-shaped structure,

for a system comprising a state of charge SOC_iValue of (OCV) with voltage_iThe approximation of the mapping of the changes is,

-determining state of health, SOH_iApproximations, which can only be estimated by means of calculations or mathematical models,

factory capacity Q_iDeviation from tolerance of (2).

In these calculations, the aforementioned inaccuracies must be taken into account, not only the quantities related to unit i, but also the quantities related to the target unit.

Calculating the amount of charge Δ q actually discharged by the cell i during the relaxation time interval T_iOther inaccuracies may occur at (T).

According to the illustrated example, the amount of charge Δ q actually discharged is given by the following equation for any cell i of the battery 10_i(t)：

In this case, the origin of the inaccuracy includes, among others:

resistance R_BalancingThe resistance may vary with temperature,

the voltage sensor measures the voltage at the terminals to the cell i.

The amount of charge Δ q actually discharged can also be obtained from the following equation for any cell i of the battery 10_i(t)：

Δq_i(T)＝I_Balancing×t_{i, balance of} (11)

In this case, the origin of the inaccuracy comprises, in particular, the balancing current I measured by the current sensor_Balancing(and is assumed to be equal for all cells) and the true balance current I flowing through cell I_{i, balance}The difference between them.

In view of the above, in the illustrated example, for any cell i, a threshold value ε is defined that combines all of the above-mentioned inaccuracies_i. To calculate the threshold value epsilon_iThe error δ q for calculating the expected charge to be balanced for cell i will be explained in detail_{i, balance}And error delta q of the amount of charge actually discharged by cell i_i(T). For any cell i of the battery 10, based on the following equation:

q_{i, balance of}＝SOC_i×Q_i-SOC_Target×Q_Target (2)

The first differential of the expected charge to be balanced for cell i can be obtained:

this equation can be simplified as follows:

dq_{i, balance of}＝Q_idSOC_i+SOC_idQ_i-Q_TargetdSOC_Target-SOC_TargetdQ_Target (20)

For any cell i of battery 10, the error δ q of the expected charge to be balanced for cell i is therefore calculated as follows_{i, balance}：

δq_{i, balance of}＝|Q_i.δSOC_i|+|SOC_i.δQ_i|+|Q_Target.δSOC_Target|+|SOC_Target.δQ_Target|

(21)

Wherein, delta SOC_iError, δ Q, representing the state of charge of cell i_iError, δ SOC, representing the capacity of cell i_TargetError representing the charge state of the target cell, and δ Q_TargetRepresenting the error in the capacity of the target cell.

For calculating the error δ q_{i, pingWeighing apparatus}For time t₁And time t₂Is effective. For example, the state of charge SOC estimated in a general manner for any cell i of the battery 10_iIs equivalent to replacing the delta SOC by 0.03_iIs present. The same applies to δ Q_iIs present.

To obtain the pair error δ q_{i, balance}Can be further analyzed using the following equation_iAnd delta SOC_i：

(4)

In this case, for any cell i of the battery 10, the state of charge SOC is calculated based on the zero-current voltage after relaxation of the cell i_iError delta SOC_iCaused by the following factors:

voltage sensor measuring zero current voltage OCV_i，

For a signal containing the state of charge SOC_iValue of (OCV) with voltage_iApproximation of a varying mapping for which the sensitivity dcv/dSOC is a function of the voltage OCV_iBut varies because the sensitivity is a non-linear function.

Error δ Δ q with respect to amount of actually discharged charge_i(T), for any cell i of the battery 10, the quantity Δ q can be obtained from the following equation_i(T)：

Based on this expression, the first differential of the amount of charge actually discharged by cell i can be obtained:

this equation can be simplified as follows:

for any cell i of battery 10, error δ Δ q of the amount of charge actually discharged by cell i is thus calculated as follows_i(T)：

Wherein, δ R_BalancingRepresents the equilibrium resistance R_BalancingAnd δ γ represents the error of the voltage at the terminal of the cell i.

The error δ γ of the voltage at the terminals of the cell i is essentially white noise. Thus, in the exemplary embodiment illustrated, the error δ γ is assumed to be equal to zero. The quantity Δ q can also be obtained from the following equation_i(T)：

Δq_i(T)＝I_Balancing×t_{i, balance} (11)

By obtaining the first order differential, we obtain:

for any cell i of battery 10, the error δ Δ q of the amount of charge actually discharged by cell i may then be calculated as follows_i(T)：

δΔq_i(T)＝|T.δI_Balancing|+|I_Balancing.δT| (26)

Wherein, δ I_BalancingIndicates the error in the balancing current and δ T indicates the error in the balancing time.

The error δ T can be generally considered to be zero because it is negligible.

In view of the above, for any cell i of the battery 10, the following may be appliedCalculating the error delta deltaq by one of two simplified equations_i(T)：

And delta q_i(T)＝|T.δI_Balancing| (28)

For any cell i of the battery 10, based on the following equation:

∑_i＝q_{i, balance}(t₂)+Δq_i(T)-q_{i, balance of}(t₁) (15)

Taking into account the above-mentioned errors, the equation for calculating the balance Σ can be written_iEquation (c):

∑_i＝(q_{i, balance of}(t₂)±δq_{i, balance}(t₂))+(Δq_i(T)±δΔq_i(T))-(q_{i, balance of}(t₁)±δq_{i, balance of}(t₁)) (29)

Therefore, equation (29) can also be written as:

∑_i＝(q_{i, balance}(t₂)+Δq_i(T)-q_{i, balance}(t₁))±(δq_{i, balance}(t₂)+δΔq_i(T)+δq_{i, balance}(t₁)) (30)

Therefore, the previous error δ q for error must be used_{i, balance}(t₁)、δq_{i, balance}(t₂) And delta q_i(T) the equation written defines the threshold ε as follows:

ε＝δq_{i, balance}(t₂)+δΔq_i(T)+δq_{i, balance}(t₁) (31)

Thus, the manner in which self-discharge defects in any cell of the battery 10 are detected has now been described, both theoretically and in practice. Referring to fig. 2, the method according to the present invention includes: calculating the charge balance Σ using equation (15)_iAnd calculating the threshold value ε by using equations (31), (21), (27), and (28)_iA fourth phase P04. Finally, the method includes balancing sigma_iWith a threshold value epsilon_iA fifth phase P05 of comparison is performed. In stage P05, if sigma is balanced for any unit i_iGreater than or equal to a threshold value epsilon_iThen the self-discharge defect of cell i is detected. In addition, during the phase P05, the foregoing test is performed to detect whether the target cell has a self-discharge defect by observing whether the condition of equation (18) is satisfied.

By using the threshold values calculated in this way, undetected self-discharge defects as well as false detection of non-existing self-discharge defects can be avoided. In particular, the threshold value ε_iMost of the inaccuracies of the units suitable for taking into account the influence on the electric storage batteries, in particular of the type of motor vehicle that incorporates electric propulsion. In this way, the reliability of the detection method according to the invention is maximized.

In general, the method according to the invention can be used to detect self-discharge defects in cells of a battery at an early stage, without the addition of any hardware devices, simply by software modification.

In particular, the method according to the invention does not require the comparison of the charge states of the cells after a longer relaxation period. The method according to the invention thus makes it possible to accelerate the detection of self-discharge defects, so that such detection can be carried out more frequently and therefore at an earlier stage.

Claims

1. A method for detecting self-discharge defects in cells (1, 2, 3, 4) in an electrical storage battery (10) having a plurality of battery cells (1, 2, 3, 4), wherein:

-balancing the charge of the battery cells (1, 2, 3, 4) at least partially,

-relaxation of the battery cells (1, 2, 3, 4),

-calculating, for each battery cell i, the charge balance Σ of said cell i during this balancing and relaxation_iAnd are each and every

-for each battery cell i, based on the charge balance Σ during the balancing and relaxation calculated for said cell i_iTo detect the cell iThe possible presence of any self-discharge defects,

wherein, for each battery cell i, the charge balance Σ of said cell i during the balancing and relaxation is calculated taking into account at least one quantity selected from_i: charge q to be balanced for the cell i immediately before balancing the cell i is started_{i, balance}(t₁) A charge q to be balanced for the unit i immediately after the relaxation of the unit i is finished_{i, balance}(t₂) And the amount of charge Δ q discharged by the cell i during this balancing_i(T)，

Wherein the charge balance sigma of said cell i during the balancing and relaxation is calculated for any battery cell i by applying the following relation_i：

∑_i＝q_{i, balance}(t₂)+Δq_i(T)-q_{i, balance}(t₁)

Wherein, "∑_i"denotes the charge balance of the unit i during the equilibrium and relaxation,

“q_{i, balance}(t) "represents the charge to be balanced for the cell i at time t,

“Δq_i(T) "represents the amount of charge that the cell i discharges during the time interval T,

“t₁"denotes the moment immediately before the start of balancing the cell i,

“t₂"denotes a time immediately after the end of relaxation of the unit i, and

"T" denotes a period during which balancing has been performed on the charges of the cell i.

2. The method of claim 1, wherein, for each cell i, the balancing comprises: calculating a charge q to balance for the cell_{i, balance of}Step (E02); and a step (E03) of performing the balancing, in which the amount of charge Deltaq discharged by the cell i is calculated_i(t); and as long as the charge q to be balanced for the cell i_{i, balance}StrictGreater than the amount of charge Δ q discharged by the cell i_i(t), the balancing step (E03) is continued.

3. The method of claim 1 or 2, wherein, for each cell i, the balancing comprises: calculating a charge q to balance for the cell i_{i, balance of}A step (2); and on the basis of the calculated charge q to be balanced for the cell i_{i, balance of}To calculate the equilibration time t of the cell i_{i, balance}Wherein at the calculated equilibrium time t_{i, balance}During which the balancing is performed.

4. A method according to claim 1 or 2, wherein for each cell i, the charge q to be balanced for the cell is calculated based on a parameter selected from_{i, balance}: state of charge SOC of the cell i_iState of charge SOC of target cell_TargetCapacity Q of the unit i_iCapacity Q of target unit_TargetState of health SOH of the cell i_iState of health SOH of target cell_TargetZero current voltage OCV of the cell i_iTarget unit zero current voltage OCV_TargetNominal capacity Q of said unit i_{i, nominal}And nominal capacity Q of the target unit_Target, _{Nominal scale}(ii) a And/or wherein, for each cell i, the amount of charge Δ q discharged by said cell i during the period T is calculated on the basis of at least one parameter chosen from_i(T): balance resistance R_BalancingVoltage v at the terminals of the cell i at a time T belonging to the period T_i(t) balancing currents I of the cells (1, 2, 3, 4) of the storage battery (10)_BalancingAnd the nominal voltage V of the cells (1, 2, 3, 4) of the storage battery (10)_{Nominal scale}。

5. The method of claim 1 or 2, wherein the possible presence of a self-discharge defect in the first battery cell is detected when: for any second battery except the first battery unitA unit i, a charge q to be balanced by the second unit i immediately after the relaxation of the second unit i is finished_{i, balance}(t₂) Is strictly greater than the charge q to be balanced by the second cell i immediately before the balancing of the second cell i is started_{i, balance}(t₁) And the amount of charge Δ q discharged by the second cell i during the balancing and relaxation_i(T) is not zero.

6. The method of claim 1 or 2, wherein for each cell i, the possible presence of a self-discharge defect in the cell i is detected if: the calculated charge balance sigma of the cell during the balancing and relaxation_iExceeding a strictly positive threshold epsilon₁Said threshold value ε₁Is determined based on at least one error selected from: error δ q of charge to be balanced for the cell i immediately before balancing the cell i is started_{i, balance of}(t₁) An error δ q of electric charges to be balanced for the unit i immediately after the relaxation of the unit i is finished_{i, balance}(t₂) And the error delta deltaq of the amount of charge discharged by the cell i during this balancing and relaxation_i(T)。

7. The method of claim 6, wherein the threshold epsilon is for any cell i_iIs based on the error deltaq of the charge to be balanced for the cell i_{i, balance}Determined by said error δ q_{i, balance of}Is calculated for any time t according to the following expression:

Wherein, the value of "δ q_{i, balance}"indicates the error in the charge to be balanced by the cell i at that time,

“δSOC_i"represents the error in the charge state of the cell i,

“δQ_i"indicates an error of the charge of the cell i,

“SOC_i"indicates the charge state of the cell i at the time t,

“Q_i"denotes the charge of the cell i at the instant t,

“δSOC_target"indicates an error in the charge state of the target cell,

“δQ_target"indicates an error in the charge of the target cell,

“SOC_target"represents the charge state of the target cell at said time t, and

“Q_target"represents the charge of the target cell at the time tmet.

8. The method of claim 6, wherein for any cell i, the threshold is based on an error δ Δ q in an amount of charge that the cell i discharges during a time interval T_i(T), said error δ Δ q_i(T) is calculated by using at least one of the following expressions:

wherein "T" represents the time interval,

“δΔq_i(T) "represents the error in the amount of charge discharged by the cell i during the time interval T,

“R_balancing"means the resistance used for this balancing,

for any time instant τ, V, belonging to said time interval_i(τ) "represents the voltage measured at the terminal of the cell i at that instant τ, and

“δI_balancing"denotes the permissible deviation of the balancing currents of the cells (1, 2, 3, 4) of the storage cell (10).